OpenAI, the creator of ChatGPT, announced on Friday that it is limiting the launch of its latest artificial intelligence model at the request of the U.S. President Donald Trump’s administration. This move is part of a growing trend where the government is scrutinizing AI products that may present cybersecurity threats.
The new AI model, named GPT-5.6 Sol, will initially only be accessible to a select group of approved partners sanctioned by the Trump administration. OpenAI emphasized that this government-mandated access process should not become the standard procedure in the long run.
While this restricted release is seen as a temporary measure by OpenAI, the company is planning to make the AI model more widely available in the upcoming weeks. This cautious approach comes after the recent actions taken by the government against OpenAI’s competitor, Anthropic, the developer of the Claude chatbot.
Anthropic voluntarily took down its latest AI models, Fable 5 and Mythos 5, shortly after their public release to comply with a directive from the Trump administration, which prohibited their use by foreign individuals. The White House continues to engage with leading AI research labs to address the challenges associated with the rapid expansion of AI technology.
OpenAI’s decision to stagger the introduction of its powerful new AI system aligns with the executive order signed by Trump in June, which established a process for assessing the national security implications of advanced AI systems before their public launch. The company clarified that its Sol model prioritizes identifying and rectifying vulnerabilities rather than engaging in cyberattacks.
Despite asserting that the Sol model does not surpass OpenAI’s risk threshold, the company acknowledged potential unforeseen risks, especially in conjunction with other tools. Consequently, OpenAI is implementing enhanced safeguards and a phased rollout strategy to mitigate these uncertainties.
A group of cybersecurity experts has criticized the government’s actions leading to the shutdown of Anthropic’s Fable model, deemed a safer alternative to Mythos. These experts argue that the concerns raised lack substantial evidence and are not unique to Fable, with similar risks existing in publicly available AI models from other sources.
Alex Stamos, a cybersecurity expert and Chief Product Officer at Corridor, highlighted that the government’s measures could hinder the U.S.’s competitiveness in the AI race against China. Stamos emphasized the importance of a balanced approach to AI oversight and innovation to maintain a strategic advantage.
In light of these developments, OpenAI continues to navigate the evolving landscape of AI regulation and security, aiming to strike a balance between innovation and risk management.